ddg-search-hardened

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill performs web searches via DuckDuckGo (described in SKILL.md and implemented by scripts/search.py using the ddgs library) and explicitly instructs the agent to use web_fetch to read full pages from result URLs, meaning it ingests public, untrusted third-party content that the agent is expected to interpret and that could materially influence actions.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly instructs installing a package with "pip install --break-system-packages ddgs", which encourages bypassing distro package protections and modifying system-wide files, so it can lead to compromising the host state.